Publications

Results 26–50 of 104

Search results

Jump to search filters

Unsupervised learning methods to perform material identification tasks on spectral computed tomography data

Proceedings of SPIE - The International Society for Optical Engineering

Gallegos, Isabel; Koundinyan, Srivathsan; Suknot, April; Jimenez, Edward S.; Thompson, Kyle; Goodner, Ryan N.

Sandia National Laboratories has developed a method that applies machine learning methods to high-energy spectral X-ray computed tomography data to identify material composition for every reconstructed voxel in the field-of-view. While initial experiments led by Koundinyan et al. demonstrated that supervised machine learning techniques perform well in identifying a variety of classes of materials, this work presents an unsupervised approach that differentiates isolated materials with highly similar properties, and can be applied on spectral computed tomography data to identify materials more accurately compared to traditional performance. Additionally, if regions of the spectrum for multiple voxels become unusable due to artifacts, this method can still reliably perform material identification. This enhanced capability can tremendously impact fields in security, industry, and medicine that leverage non-destructive evaluation for detection, verification, and validation applications.

More Details
TYPE Conference Poster YEAR 2018
DOIOSTIScopus

A kinetic approach to modeling the manufacture of high density strucutral foam: Foaming and polymerization

Rao, Rekha R.; Mondy, Lisa A.; Noble, David R.; Brunini, Victor; Roberts, Christine; Long, Kevin N.; Soehnel, Melissa; Celina, Mathew C.; Wyatt, Nicholas B.; Thompson, Kyle

We are studying PMDI polyurethane with a fast catalyst, such that filling and polymerization occur simultaneously. The foam is over-packed to tw ice or more of its free rise density to reach the density of interest. Our approach is to co mbine model development closely with experiments to discover new physics, to parameterize models and to validate the models once they have been developed. The model must be able to repres ent the expansion, filling, curing, and final foam properties. PMDI is chemically blown foam, wh ere carbon dioxide is pr oduced via the reaction of water and isocyanate. The isocyanate also re acts with polyol in a competing reaction, which produces the polymer. A new kinetic model is developed and implemented, which follows a simplified mathematical formalism that decouple s these two reactions. The model predicts the polymerization reaction via condensation chemis try, where vitrification and glass transition temperature evolution must be included to correctly predict this quantity. The foam gas generation kinetics are determined by tracking the molar concentration of both water and carbon dioxide. Understanding the therma l history and loads on the foam due to exothermicity and oven heating is very important to the results, since the kinetics and ma terial properties are all very sensitive to temperature. The conservation eq uations, including the e quations of motion, an energy balance, and thr ee rate equations are solved via a stabilized finite element method. We assume generalized-Newtonian rheology that is dependent on the cure, gas fraction, and temperature. The conservation equations are comb ined with a level set method to determine the location of the free surface over time. Results from the model are compared to experimental flow visualization data and post-te st CT data for the density. Seve ral geometries are investigated including a mock encapsulation part, two configur ations of a mock stru ctural part, and a bar geometry to specifically test the density model. We have found that the model predicts both average density and filling profiles well. However, it under predicts density gradients, especially in the gravity direction. Thoughts on m odel improvements are also discussed.

More Details
TYPE SAND Report YEAR 2015
DOIOSTI
Results 26–50 of 104
Results 26–50 of 104